999

Journal of Biomolecular Structure & Dynamics, ISSN 0739-1102 Volume 29, Issue Number 5, (2012) ©Adenine Press (2012)

*Corresponding authors:Dr. Jinhyuk LeeDr. Guo-Ying QianPhone: 82-428798530Fax: 82-428798519E-mail: jinhyuk@kribb.re.krPhone: 86-574-88222298Fax: 86-574-88222298E-mail: qianguoying_wanli@hotmail.com

Yue-Xiu Si1

Shang-Jun Yin1

Sangho Oh2

Zhi-Jiang Wang1

Sen Ye1 Li Yan3

Jun-Mo Yang4

Yong-Doo Park1,3

Jinhyuk Lee2,5*Guo-Ying Qian1*

1College of Biological and

Environmental Sciences, Zhejiang Wanli

University, Ningbo 315100, P. R. China2Korean Bioinformation Center

(KOBIC), Korea Research Institute of

Bioscience and Biotechnology, Daejeon

305-806, Korea3Zhejiang Provincial Key Laboratory

of Applied Enzymology, Yangtze Delta

Region Institute of Tsinghua University,

Jiaxing 314006, P. R. China4Department of Dermatology,

Sungkyunkwan University School of

Medicine, Samsung Medical Center,

Seoul 135-710, Korea5Department of Bioinformatics,

University of Sciences and Technology,

Daejeon 305-350, Korea

An Integrated Study of Tyrosinase Inhibition by Rutin: Progress using a Computational Simulation

http:::www.jbsdonline.com

Abstract

Tyrosinase inhibition studies have recently gained the attention of researchers due to their potential application values. We simulated docking (binding energies for AutoDock Vina: 29.1 kcal:mol) and performed a molecular dynamics simulation to verify docking results between tyrosinase and rutin. The docking results suggest that rutin mostly interacts with histidine residues located in the active site. A 10 ns molecular dynamics simulation showed that one copper ion at the tyrosinase active site was responsible for the interaction with rutin. Kinetic analyses showed that rutin-mediated inactivation followed a first-order reaction and mono- and biphasic rate constants occurred with rutin. The inhibition was a typical competi-tive type with Ki 5 1.10 6 0.25 mM. Measurements of intrinsic and ANS-binding fluores-cences showed that rutin showed a relatively strong binding affinity for tyrosinase and one possible binding site that could be a copper was detected accompanying with a hydrophobic exposure of tyrosinase. Cell viability testing with rutin in HaCaT keratinocytes showed that no toxic effects were produced. Taken together, rutin has the potential to be a potent anti-pigment agent. The strategy of predicting tyrosinase inhibition based on hydroxyl group number and computational simulation may prove useful for the screening of potential tyro-sinase inhibitors.

Key words: Tyrosinase; Inhibition kinetics; Rutin; Hydroxyl group; Docking simulation.

Introduction

Tyrosinase (EC 1.14.18.1) is a ubiquitous enzyme with diverse physiological roles related to pigment production. Tyrosinase catalyzes pigmentation of the skin (1, 2), the browning of vegetables (3, 4), wound healing (5, 6), and cuticle formation in insects (7, 8). Tyrosinase inhibitors have potential applications in medicine, in cosmetics as whitening agents, and in agriculture as bio-insecticides. Tyrosinase belongs to the type 3 copper protein family (9, 10), with two copper ions each coordinately bonded to a distinct set of three histidine residues within the active site. These coppers participate directly in the hydroxylation of monophenols to o-diphenols (cresolase activity) and in the oxidation of o-diphenols to o-quinones (catechol oxidase activity) (11). The tyrosinase mechanism is complex in that this enzyme can catalyze multiple reactions. The crystallographic structure of tyrosi-nase has recently been determined (12); as such, the overall 3D structural changes that accompany activity modulations with various ligands can now be well under-stood. Studies of this enzyme mechanism must involve a variety of kinetic and computational methods to derive structure-function relationships.

Abbreviations used: DOPA: 3,4-dihydroxyphenylalanine; ANS: 1-anilinonaphthalene-8-sulfonate; MTS: 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium.

Open Access ArticleThe authors, the publisher, and the right holders grant the right to use, reproduce, and disseminate the work in digital form to all users.

1000

Si et al.

Rutin, also known as quercetin-3-rutinoside, is naturally found in vegetables and fruits (13-16). Rutin consists of a flavonoid quercetin and the disaccharide ruti-nose via glycoside binding; therefore, bonding of disaccharide onto the quercetin hydroxyl group produces it on a large scale. Rutin is used in numerous multivitamin preparations and herbal remedies (17, 18). The biomedical applications of rutin have been reported: role as an antioxidant (19, 20) and its use in the treatment of diabetes (21, 22). Rutin has been shown to have anti-inflammatory (23), anti-cancer (24, 25), anti-angiogenesis (26) activities.

In this study, we investigated the mechanism of tyrosinase inhibition by rutin using kinetic analyses and computational simulations. We hypothesized that the ten rutin hydroxyl groups may block L-DOPA oxidation by binding to tyrosinase. Previous findings have shown the importance of hydroxyl groups in tyrosinase inhibition (27-33) in terms of molecular position, number, and specific interactions with the enzyme, further supporting our hypothesis. Due to the fact that the next step of functional studies for rutin is a time- and cost-consuming process and for the right choice, we adapted the computational predictions on the basis of our hypothesis at the initial step. Computational simulations as well as kinetic studies confirmed the inhibitory action of rutin on tyrosinase. The kinetic parameters suggested that rutin binds to the active site residues and, as a result, induces copper chelation and hydrophobic exposure. Our study suggests new insights into tyrosinase inhibition and potential applications.

Materials and Methods

Materials

Tyrosinase (M.W. 128 kDa), L-DOPA, and rutin hydrate were purchased from Sigma-Aldrich. When L-DOPA was used as a substrate in our experiments, the pur-chased tyrosinase had a Km of 0.71 6 0.04 mM (Vmax 5 0.17 6 0.005 mmol min21) according to a Lineweaver-Burk plot. DMSO (5%) was used for dissolving rutin into solution.

Tyrosinase Assay

A spectrophotometric tyrosinase assay was performed as previously described (34, 35) with a modification to the buffer condition (50 mM Tris-HCl, pH 8.8). To begin the assay, a 10-ml sample of enzyme solution was added to 1 ml of reaction mix. Tyrosinase activity (v) was recorded as the change in absorbance per min at 492 nm using a Perkin Elmer Lambda Bio U:V spectrophotometer.

In silico Docking of Tyrosinase and Rutin and Molecular Dynamics Simulation

We used the crystal structure of Agaricus bisporus tyrosinase (PDB ID: 2Y9X) to simulate the tyrosinase structure (12). Among the many tools available for in silico protein-ligand docking, AutoDock Vina (36) was used because of its automated docking capability. The program performs ligand docking using a set of predefined 3D grids of the target protein. The original structure of rutin was derived from the PubChem database (Compound ID: 5280805) (http:::pubchem.ncbi.nlm.nih.gov). The following steps were performed in preparation of the docking procedure: 1) conversion of 2D structures to 3D structures, 2) calculation of charges, 3) addi-tion of hydrogen atoms, and 4) localization of pockets. For these steps, we used the OMEGA 2.0 OpenEye package. To verify our docking results, we performed a 10 ns production molecular dynamics simulation using CHARMM (37). Initial structure for the simulation was generated using “PDB reader” in the CHARMM-GUI website (38) for tyrosinase and CGENFF (39) for rutin. A generalized Born model with a simple switch function (GBSW) (40) was used to consider the solvation effect. The

1001

Tyrosinase Inhibition by Rutin

structures were saved every 1 ps for trajectory analysis. We measured the structural details of the interactions as a function of time to ensure that the interactions revealed by the docking study were conserved during the simulations.

Kinetic Analysis for Competitive-Type Inhibition

To describe the competitive inhibition mechanism, the Lineweaver-Burk equation in double reciprocal form can be written as:

11

1 1

v

K

V K Vm

i

= + [ ]

[ ] +

max max

I

S [1]

and secondary plots can be constructed from

KK

KKm

m

im

app I=

[ ]+ [2]

As such, the Ki, Km, and Vmax values can be derived from the above equations. The secondary plot of the apparent Km vs. [I] was linearly fitted, assuming a single inhi-bition site or a single class of inhibition site.

Intrinsic and ANS-Binding Fluorescence Measurements

Fluorescence emission spectra were measured using a Jasco FP750 spectrofluo-rometer with a 1-cm path length cuvette. Tryptophan fluorescence was measured following excitation at 280 nm, and the emission wavelength ranged between 300 and 410 nm. Changes in the ANS-binding fluorescence of tyrosinase were mea-sured following excitation at 390 nm, and the emission wavelength ranged from 400 to 520 nm. The tyrosinase was labeled with 40 mM ANS for 30 min prior to measurements.

All kinetic reactions and measurements in this study were performed in 50 mM Tris-HCl buffer (pH 8.8).

Determination of the Binding Constant and the Number of Binding Sites

According to a previous report (41), when small molecules are bound to equivalent sites on a macromolecule the equilibrium between the free and bound molecules is given by the following equation:

F

F F n K Q0

0

1 1 1

=

[ ] [3]

where F0 and F are the relative steady-state fluorescence intensities in the absence and presence of quencher, respectively, and [Q] is the quencher (rutin) concentra-tion. The values for the binding constant (K) and number of binding sites (n) can be derived from the intercept and slope of a plot based on Eq. [3].

HaCaT Cell Culture and MTS Assay

The HaCaT cell line consists of immortalized normal keratinocytes (42). This cell line was cultured in DMEM containing 10% FBS and 1% antibiotic-antimycotic (Gibco, Rockville, MD). The cytotoxicity of rutin on HaCaT cells was tested using an MTS assay kit purchased from Sigma-Aldrich according to the manufacturer’s instructions.

1002

Si et al.

Results

Rutin Docking and Molecular Dynamics Simulations

Because the crystallographic structure of tyrosinase from Agaricus bisporus is now available (PDB ID: 2Y9X) (12), we constructed the 3D structure of tyrosinase using MODELLER (43). For the docking programs, we used a set of 3D grids of the tar-get protein with a systematic search technique. In the 3D structure of tyrosinase, the docking simulation of binding between rutin and tyrosinase was successful in pro-

ducing a binding energy score of 29.1 kcal:mol for AutoDock Vina (Figure 1A). We searched for common neighboring rutin-binding res-idues within tyrosinase and found several residues including HIS56, TYR73, LYS74, ALA75, HIS80, GLY81, PHE85, HIS89, TYR179, HIS234, ASP238, GLU246, HIS249, ASP250, HIS253, ASP266, HIS267, PRO268, PHE269, HIS281, ARG306, and ASP307. For the control simulation, ascorbic acid was also docked with tyrosinase since ascorbic acid has functions of antioxidant as well as tyrosinase inhibition, which is similar to rutin (Figure 1B). The binding energy of ascorbic acid was 25.4 kcal:mol. The docking scores indicate that rutin binds more tightly to tyrosinase than ascorbic acid. We probed several binding residues for ascorbic acid such as HIS249, ASP251, VAL254, HIS264, MET265, ASP266, and HIS267. Compared to the result of rutin, HIS249, HIS267 and ASP266 were commonly shared the interactions for the two different inhibitors.

To see structural changes during simulation, we performed a 10 ns molecular dynamics simulation (Figure 2). We verified the root mean square deviation (RMSD) of the alpha carbon with a reference (the AutoDock Vina structure), and the results showed that the structure was first rearranged and then stabilized (Figure 2A). Using the final structure, we chose candidate residues for possible interactions with rutin. We measured all of distances of these candidates via simu-lation. As shown in Figure 2B and C, we suggested four plausible interactions from the four main aromatic rings (labeled with 1 to 4) of rutin. The reason that we divided rutin into the four rings was because it has flexible rings during the simulation, making it difficult to measure their specific interactions with tyrosinase. The interac-tions gradually stabilized after 7.5 ns as the simulation progressed (Figure 2C). The plausible residues determined by the molecular dynamics simulations included the following: ALA75, HIS267, PRO268, PHE269, and ASP307 by Ring 1; HIS56, HIS80, PHE85, PRO268, and ASP307 by Ring 2; ARG306 and ASP307 by Ring 3; and GLU246, HIS249, and ASP250 by Ring 4. Ring 2 was accessed as the core site of the pocket and was stabilized. Ring 3 showed a relatively weak stability with a fewer number of plausible residues and larger fluctuations in the distances since it is located at the sur-face of the active site pocket after positioning. In our approach, the identified plausible residues for docking and the molecular dynamics simulations were consistent.

We determined the time profile for the plausible binding of two cop-pers with rutin (Figure 3A). As a result, the most conspicuous bind-ing with respect to stability, distance, and fixation occurred at Ring 2 of rutin where the reaction stabilized after 7.5 ns. Ring 2 was closest to the coppers, and between the two coppers, copper B (lower part) was more tightly bound to the hydroxyl group of Ring 2 (Figure 3B). Compared to the results of Figure 2, Ring 2 of rutin was thought to bind to both copper ions at the active site; however, only one copper

Figure 1: Computational simulations between tyrosinase and rutin. (A) The protein structure is taken from the final molecular dynamics simulation. Rutin was docked in the pocket site. AutoDock Vina pre-dicted magenta residues, and cyan residues were obtained from CHARMM. The two blue spheres indicate the copper ions (copper A, upper side; copper B, lower side). (B) The comparative docking simula-tions. Magenta represents ascorbic acid used as a control inhibitor as docked by AutoDock Vina. All conditions were as for (A).

1003

Tyrosinase Inhibition by Rutin

was stabilized in the reaction. This may be due to the preferable binding between Ring 2 and copper B by structural conformation and direction. After the occurrence of strong binding in Ring 2, the other residues around Ring 2 of rutin within the active site pocket were relatively stabilized.

Inhibition Kinetics: Effect of Rutin on the Activity of Tyrosinase

Tyrosinase activity was conspicuously inactivated by rutin in a complex man-ner with an IC50 of 6.8 6 0.3 mM (n 5 3) (Figure 4A). At low rutin concentra-tions (less than 2.0 mM), we repeatedly observed that tyrosinase was slightly

Figure 2: Molecular dynamics simulations between tyrosinase and rutin. (A) RMSD time profiles of carbon alpha with the reference structure (AutoDock Vina structure). (B) Four main aromatic rings (Rings 1 to 4). (C) Time profiles of plausible interactions with the four main aromatic rings from rutin. The distances were measured between the two geometric centers of its side-chain and of the aromatic ring from rutin.

1004

Si et al.

activated by 6 to 8%. At 20 mM rutin, tyrosinase was completely inactivated. When rutin was absent from the assay (it caused dilution effect), the IC50 value shifted to 12 6 2.83 mM (n 5 3), indicating that rutin reversibly inhibited tyrosinase (Figure 4B). The slight activation at low rutin concentration was also observed in this condition. To confirm the reversibility of rutin-mediated inhibition, a plot of the remaining activity versus [E] was constructed (Figure 5A). As expected, the results showed straight lines passing through the origin, indicating that the inhibition

Figure 3: Strong binding interaction between copper within the active site and rutin. (A) Time profiles of the two copper distances, copper A (green) and copper B (red), with the four main aromatic rings of rutin. (B) Close-up view of the active site. Interaction between copper B and the oxygen in Ring 2 is drawn as a red solid line.

1005

Tyrosinase Inhibition by Rutin

by rutin was reversible. Rutin could directly chelate a copper ion at the active site of tyrosinase due to the binding property, which was directly suggested by the UV-scanning spectra (Figure 5B).

Next, the kinetic time-courses for tyrosinase at different concentrations of rutin were compared to determine inactivation rate constants as well as to detect whether an unfolding process was involved (Figure 6). The results showed that the cata-lytic rate changed detectably with time from lower (3.75 mM) to higher (8.75 mM) rutin concentrations with a first-order reaction. At the highest concentration condi-tion (20 mM), rutin appeared to inactivate tyrosinase very quickly and to induce no further kinetic change in the activity. Interestingly, subsequent analyses from the semi-logarithmic plots (Figure 7) showed mono- and multi-phasic inactiva-tion processes. When the rutin concentration was 3.75 mM, tyrosinase inactivation followed a monophasic process and the microscopic rate constant was calculated as k 5 0.36 6 0.03 3 1023 s21 (Figure 7A). However, when rutin concentra-tion was increased to 6.25 mM, the inactivation process split into fast and slow

0.0 4.0 8.0 12.0 16.0 20.00

40

80

120

20

60

100

Rel

ativ

e ac

tivi

ty (

%)

Rel

ativ

e ac

tivi

ty (

%)

0.0 10.0 20.0 30.0

[Rutin] (mM)

[Rutin] (mM)

0

20

40

60

80

100

120

(A)

(B)

Figure 4: Inhibitory effect of rutin on tyrosinase. Data are presented as mean (n 5 3). Tyrosinase was incubated with rutin at various concentrations for 3 h at 25ºC and then added to the assay system at the corresponding rutin concentration (A) or in the absence of rutin (B). The final concentrations of L-DOPA and tyrosinase were 2 mM and 2.0 mg:ml, respectively.

1006

Si et al.

processes (Figure 7B), implying that tyrosinase transiently passes through intermediates until it reaches a completely inactive state. The micro-scopic rate constants were calculated as kfast 5 2.43 6 0.43 3 1023 s21 and kslow 5 0.16 6 0.04 3 1023 s21, respectively. The rate constants for 8.75 mM rutin were also calculated as kfast 5 1.46 6 0.2 3 1023s21 and kslow 5 0.18 6 0.03 3 1023 s21, respectively. They all ranged within the same order of magnitude.

Competitive Inhibition by Rutin Detected by Lineweaver-Burk Analysis

Lineweaver-Burk plots showed no changes in the Km or apparent Vmax, indicating that rutin induced a typical competitive inhibition (Figure 8A). The secondary plot of Km versus [Rutin] was linearly fitted (Figure 8B), showing that rutin has a single inhibition site or a single class of inhibi-tion sites on tyrosinase. Using Eq. [2], the Ki value was calculated as 1.10 6 0.25 mM (n 5 3). This result directly indicated that rutin com-peted with L-DOPA substrate in the active site pocket due to the prop-erty of copper chelating. Since rutin has hydroxyl functional groups, the strong copper chelating property was expected.

0.0 1.0 2.0 3.0 4.0[E] mg/ml

0.00

0.05

0.10

0.15

0.20

0.25

v (D

OD

/min

)

1

2

3

4

5

300 400 500 600

Wavelength (nm)

0.0

0.4

0.8

1.2

1.6

2.0

Abso

rban

ce

Rutin + Cu2+

Cu2+

Rutin

(A)

(B)

Figure 5: Rutin binding to tyrosinase and copper ion. (A) Plots of v vs. [E]. The v value indicates the change in absorbance at 492 nm per min at rutin concentrations of labels 1 to 5 as 0, 2.5, 5.0, 7.5, and 10 mM, respectively. The final L-DOPA concentra-tion was 2 mM. (B) UV:Visible scanning using 0.05 mM rutin and 0.2 mM CuSO4.

0 40 80 120 160 200Time (min)

0

20

40

60

80

100

Rel

ativ

e ac

tivi

ty (

%)

Figure 6: Time course of tyrosinase inhibition in the presence of rutin. The enzyme solution was mixed with rutin at 3.75 (), 6.25 (), 8.75 (), and 20 (+) mM, and aliquots were collected for assay at the indicated time intervals.

1007

Tyrosinase Inhibition by Rutin

Effect of Rutin on Tyrosinase Tertiary Structure: Spectrofluorimetry Studies

Tertiary structural changes in tyrosinase in the presence of rutin were determined from measurements of intrinsic and ANS-binding fluorescence. We found that rutin had a quenching effect on intrinsic fluorescence, which gradually decreased with no shift in the maximum peak wavelength (Figure 9A). At less than 0.4 mM, rutin completely quenched the fluorescence. A double reciprocal plot between maximum fluorescence intensity and rutin concentration revealed a linear rela-tionship (Figure 9B). From these data, we calculated the binding constant as K 5 15.27 6 0.13 mM21 and the binding number as n 5 1.18 6 0.01 according to Eq. [3]. Thus, rutin showed a relatively strong binding affinity for tyrosinase in the absence of substrate and one possible binding site on tyrosinase that could be a copper. The results from the computational simulations of molecular dynamics in Figure 3 also support the experimental result that, of the two coppers, only one copper strongly interacts with rutin at the equilibrium state.

0 20 40 60 80 1000.01

0.10

1.00

(At-A

∞)/(

A0

-A∞)

0 40 80 120Time (min)

Time (min)

0.01

0.10

1.00

(At-A

∞)/(

A0

-A∞)

(A)

(B)

Figure 7: Semi-logarithmic plot analyses. (A) Experimental points () were obtained in the presence of 3.75 mM rutin. (B) Experimental points () were obtained in the presence of 6.25 mM rutin. () Points obtained by subtracting the contribution of the slow phase (---) from the data in the curve. Slopes of the curves (— or ---) indicate the rate constants. The final concentrations of L-DOPA and the enzyme were 2 mM and 4 mg:ml, respectively.

1008

Si et al.

Next, we monitored changes in tyrosinase hydrophobicity in the presence of rutin (Figure 10). The native state of tyrosinase showed a low degree of hydropho-bicity, which was mainly induced by the active site. Interestingly, we repeat-edly observed that the relatively low concentration of rutin (0.15 mM) induced a conspicuous spectral change, implying that regional hydrophobic surfaces within tyrosinase were exposed. This phenomenon did not occur in a dose-dependent manner since 0.078 and 0.31 mM rutin did not significantly alter ANS-binding fluorescence of tyrosinase. Low rutin concentrations may loosen the active site pocket, which might result in the activation of tyrosinase by low concentrations of rutin, as observed in Figure 4 where 0.15 mM rutin enhanced the tyrosinase activ-ity by the maximum 10%.

Rutin Cytotoxicity Testing on the HaCaT Cell Line Using an MTS Assay

We applied rutin to HaCaT cells to check for the cytotoxicity of rutin on skin cells. Rutin did not induce significant toxic effects in HaCaT cells (Figure 11). This result

–2.0 0.0 2.0 4.0

1/[L-DOPA (mM)]

20

40

60

80

1/v

(DO

D/m

in)

0.0 1.0 2.0 3.0 4.0 5.0[Rutin] (mM)

0.0

1.0

2.0

3.0

4.0

(A)

(B)

Figure 8: Lineweaver-Burk plot. (A) The rutin concentrations were 0 (), 1.25 (), 2.5 (), and 5 mM (). The final enzyme concentration was 2.0 mg:ml. (B) The secondary plot. Concentrations were determined using Eq. [2].

1009

Tyrosinase Inhibition by Rutin

implies that rutin could be used as a potent tyrosinase inhibitor as a whitening agent for pigment disorders such as hyperpigmentation.

Discussion

In an earlier report, rutin and copper complexes were characterized as inhibitory factors on lipid peroxidation in rat liver microsomes (44). Direct evidence of the copper chelating property of rutin was previously suggested by electrospray ioniza-tion tandem mass spectrometry where four kinds of complexes with four different stoichiometric ratios were produced in the reaction between copper and rutin (45). Previous studies have recognized potent inhibitory effects of compounds with

0 0.1 0.2 0.3 0.4[Rutin] (mM)

[Rutin] (mM)–1

0

20

40

60

80

100

Rel

ativ

e fl

uo

resc

ence

inte

nsi

ty (

%)

0 40 80 12020 60 1000.0

2.0

4.0

6.0

8.0

F 0 /(

F 0 –

F)

(A)

(B)

Figure 9: Changes in intrinsic tyrosinase fluorescence at different rutin concentrations. (A) Maximum intrinsic fluorescence intensity changes. Tyrosinase was incubated with rutin for 3 h before measure-ment. (B) Double reciprocal plot of F0 /(F02F) vs. [Q]21. F0, maximum native fluorescence intensity; F, maximum fluorescence intensity of sample; Q, quencher rutin. The final enzyme concentration was 66 mg:ml.

1010

Si et al.

hydroxyl groups on tyrosinase (27-33). In this context, we determined that rutin, with phenolic groups, also inhibits the enzyme. We found that the inhibitory mech-anism of rutin is similar to those of copper chelators (46-50). This is evident in that rutin competes with substrate L-DOPA in the active site pocket in a competi-tive manner and induces hydrophobic surface exposure. The competitive inhibition implied that the rutin binding site somehow overlaps with the L-DOPA binding

site that is in close proximity, which may allow rutin binding to com-pete with substrate binding. Using computational simulations, we pre-dicted that rutin could bind tightly to the copper ion with several amino acid residues close to the active site of tyrosinase. These residues are thought to be involved in the first stage of rutin binding and in direct-ing the docking of copper ion. The computational simulation provided supportive data for the competitive inhibition of rutin by identifying binding residues in proximity to the binding pocket. In addition, the experimental results verified the simulations.

The slight activation in the presence of low rutin concentrations sug-gested that the shape of the tyrosinase active site was modulated to be preferable for L-DOPA accession and docking, which might be associated with the hydrophobic surface exposure. Shift of the mono-phasic inactivation process to the biphasic process (kfast and kslow) in the presence of high rutin concentrations implies that tyrosinase tran-siently passes through several distinct intermediates until it reaches a completely inactive state. Thus, the accumulation of intermediates as a result of increasing rutin concentration was likely responsible for the overall decrease in enzymatic activity. Rutin docking to the enzyme occurred very quickly but only gradually induced the loss of activity, after which it transitioned to a slow phase. Similar results have also been reported in a previous study where tyrosinase was inactivated by denaturants including urea, guanidine hydrochloride, and sodium dodecyl sulfate in a biphasic process (51, 52). The accumulation of intermediates indicates that tyrosinase has a stable tertiary structure.

Results from the intrinsic fluorescence spectra showed a possible binding site for rutin at the active site of tyrosinase, supporting the fact that, of the two coppers at the active site, only one coordinates with the three histidine residues directly che-lated by rutin. The coordinating residues guide rutin at the initial stage, and then

the copper is gradually chelated by rutin accompanying enzymatic inac-tivation. It therefore appears as one binding without L-DOPA substrate during the measurement of intrinsic fluorescence.

The principal findings of our study include the following: i) rutin binding to tyrosinase causes a competitive type of inhibition; ii) rutin inhibition of tyrosinase involves hydrophobic surface exposures at certain rutin concentrations; iii) the mono- and bi-phasic kinetic processes occurred during rutin-mediated tyrosinase inactivation; iv) as predicted by compu-tational simulations, putative rutin-binding residues were located within the active site pocket, and these residues affected the docking of rutin at the initial stage; v) one copper that coordinated with histidine residues in the tyrosinase active site was directly chelated by rutin at the equilibrium state; and vi) further study of rutin as a tyrosinase inhibitor may lead to the development of new agents for skin pigmentation disorders.

Our study provides new insight into the role of active site residues in tyro-sinase catalysis and provides useful information regarding the structural changes of tyrosinase. A combination of inhibition kinetics and compu-tational modeling may facilitate the testing of potential tyrosinase inhibi-tors, including rutin, and the prediction of their inhibitory mechanisms.

400 440 480 520 560 600 640Wavelength (nm)

0

20

40

60

80

AN

S-f

luo

resc

ence

(A

.U.)

12

3

4

Figure 10: Changes in ANS-binding fluorescence of tyrosinase at different rutin concentrations. 40 mM ANS (curve ---) was incubated with tyrosinase for 30 min to label the hydrophobic enzyme surfaces prior to fluorescence measurement. Curve 1 indicates the native state of tyrosinase, and curves 2 to 4 indicate tyrosinase incubated with 0.078, 0.15, and 0.31 mM rutin, respectively. The final enzyme con-centration was 66 mg:ml.

2 41 3 50

40

80

120

20

60

100

Rel

ativ

e ce

ll vi

abilit

y (%

)

Figure 11: Cell viability of human HaCaT keratinocytes in the presence of rutin. The MTS assay was conducted after incubating HaCaT cells cultured with various concentrations of rutin. Labels 1 to 5 indicate 0, 50, 100, 500, 1000 mM rutin, respectively.

1011

Tyrosinase Inhibition by Rutin

Acknowledgements

This study was supported by the Zhejiang Provincial Top Key Discipline of Mod-ern Microbiology and Application. Dr. Guo-Ying Qian was supported by the grant of the National Basic Research Program of China (973 Pre-research Program) (2011CB111513). Yue-Xiu Si was financially supported by the Natural Science Foundation of Ningbo City (2011A610039). Dr. Jun-Mo Yang was supported by a grant of the Korea Health 21 R&D Project (Ministry of Health, Welfare and Family Affairs, Republic of Korea, 01-PJ3-PG6-01GN12-0001) and a grant from Samsung Biomedical Research Institute (C-A6-216-3). Dr. Jinhyuk Lee was supported by a grant from Korea Research Institute of Bioscience and Biotechnology (KRIBB) Research Initiative Program. Dr. Yong-Doo Park was supported by the Natural Science Foundation of Ningbo City (2011A610019).

References

C. Olivares and F. Solano. 1. Pigment Cell Melanoma Res 22, 750-760 (2009).K. Jimbow, J. S. Park, F. Kato, K. Hirosaki, K. Toyofuku, C. Hua, and T. Yamashita. 2. Pigment Cell Res 13, 222-229 (2000).Y. J. Kim and H. Uyama. 3. Cell Mol Life Sci 62, 1707-1723 (2005).A. Rescigno, F. Sollai, B. Pisu, A. Rinaldi, and E. Sanjust. 4. J Enzyme Inhib Med Chem 17, 207-218 (2002).M. R. Kanost, H. Jiang, and X. Q. 5. Yu. Immunol Rev 198, 97-105 (2004).S. C. Lai, C. C. Chen, and R. F. Hou. 6. J Med Entomol 39, 266-274 (2002).A. Guerrero and G.. Rosell. 7. Curr Med Chem 12, 461-469 (2005).M. J. Gorman, C. An, and M. R. Kanost. 8. Insect Biochem Mol Biol 37, 1327-37 (2007).J. Yoon, S. Fujii, and E. I. Solomon. 9. Proc Natl Acad Sci USA 106, 6585-6590 (2009).Y. Li, Y. Wang, H. Jiang, and J. Deng. 10. Proc Natl Acad Sci USA 106, 17002-17006 (2009).H. Decker and F. Tuczek. 11. Trends Biochem Sci 25, 392-397 (2000).W. T. Ismaya, H. J. Rozeboom, A. Weijn, J. J. Mes, F. Fusetti, H. J. Wichers, and 12. B. W. Dijkstra. Biochemistry 50, 5477-5486 (2011).Z. Xie, Y. Zhao, P. Chen, P. Jing, J. Yue, and L. L.Yu. 13. J Agric Food Chem 59, 3042-3049 (2011).R. Slimestad and M.Verheul. 14. J Agric Food Chem 59, 3180-3185 (2011).S. Tong, J. Yan, G. Chen, and J. Lou. 15. J Chromatogr Sci 47, 341-344 (2009).J. Kalinova, J. Triska, and N. Vrchotova. 16. J Agric Food Chem 54, 5330-5335 (2006).I. Erlund, T. Kosonen, G. Alfthan, J. Mäenpää, K. Perttunen, J. Kenraali, J. Parantainen, and 17. A. Aro. Eur J Clin Pharmacol 56, 545-553 (2000).W. Dimpfel. 18. Phytomedicine 16, 287-294 (2009).S. P. Boyle, V. L. Dobson, S. J. Duthie, D. C. Hinselwood, J. A. Kyle, and A. R. Collins. 19. Eur J Clin Nutr 54, 774-782 (2000).C. La Casa, I. Villegas, C. Alarcón de la Lastra, V. Motilva, and M. J. Martín Calero. 20. J Ethnopharmacol 71, 45-53 (2000).H. D. Je, C. Y. Shin, S. Y. Park, S. H. Yim, C. Kum, I. H. Huh, J. H. Kim, and U. D. Sohn. 21. Arch Pharm Res 25, 184-190 (2002).A. A. Fernandes, E. L. Novelli, K. Okoshi, M. P. Okoshi, B. P. Di Muzio, J. F. Guimarães, 22. and A. Fernandes, J. Biomed Pharmacother 64, 214-219 (2010).K. Nones, Y. E. Dommels, S. Martell, C. Butts, W. C. McNabb, Z. A. Park, S. Zhu, 23. D. Hedderley, M. P. Barnett, and N. C. Roy. Br J Nutr 101, 169-181 (2009).E. Bourogaa, J. Bertrand, M. Despeaux, R. Jarraya, N. Fabre, L. Payrastre, C. Demur, 24. J. J. Fournié, M. Damak, A. E. Feki, and C. Leuk. Res 35, 1093-1101 (2011).J. P. Lin, J. S. Yang, J. J. Lin, K. C. Lai, H. F. Lu, C. Y. Ma, R. Sai-Chuen Wu, K. C. Wu, F. 25. S. Chueh, W. Gibson Wood, and J. G. Chung. Environ Toxicol Jan 20 (2011). doi: 10.1002:tox.20662. [Epub ahead of print]H. Luo, B. H. Jiang, S. M. King, and Y. C. Chen. 26. Nutr Cancer 60, 800-809 (2008).Q. Yan, R. Cao, W. Yi, L. Yu, Z. Chen, L. Ma, and H. Song. 27. Bioorg Med Chem Lett 19, 4055-4058 (2009).Y. Yamazaki, Y. Kawano, A. Yamanaka, and S. Maruyama. 28. Bioorg Med Chem Lett 19, 4178-4182 (2009).S. R. Kanade, V. L. Suhas, N. Chandra, and L. R. Gowda. 29. FEBS J 274, 4177-4187 (2007).D. Kim, J. Park, J. Kim, C. Han, J. Yoon, N. Kim, J. Seo, and C. Lee. 30. J Agric Food Chem 54, 935-941 (2006).M. Shiino, Y. Watanabe, and K. Umezawa. 31. Bioorg Chem 31, 129-135 (2003).M. Shiino, Y. Watanabe, and K. Umezawa. 32. Bioorg Med Chem 9, 1233-1240 (2001).T. Yokota, H. Nishio, Y. Kubota, and M. Mizoguchi. 33. Pigment Cell Res 11, 355-361 (1998).

1012

Si et al.

Z. R. Lü, E. Seo, L. Yan, S. J. Yin, Y. X. Si, G. Y. Qian, Y. D. Park, and J. M. Yang. 34. J Biomol Struct Dyn 28, 259-276 (2010).H. Y. Han, J. R. Lee, W. A. Xu, M. J. Hahn, J. M. Yang, and Y. D. Park. 35. J Biomol Struct Dyn 25, 165-171 (2007).O. Trott and A. J. Olson. 36. J Comput Chem. 31, 455-461 (2010).B. R. Brooks, C. L. 3rd Brooks, A. D. Mackerell, Jr., L. Nilsson, R. J. Petrella, B. Roux, 37. Y. Won, G. Archontis, C. Bartels, S. Boresch, A. Caflisch, L. Caves, Q. Cui, A. R. Dinner, M. Feig, S. Fischer, J. Gao, M. Hodoscek, W. Im, K. Kuczera, T. Lazaridis, J. Ma, V. Ovchinnikov, E. Paci, R. W. Pastor, C. B. Post, J. Z. Pu, M. Schaefer, B. Tidor, R. M. Venable, H. L. Woodcock, X. Wu, W. Yang, D. M. York, and M. Karplus. J Comput Chem 30, 1545-1614 (2009).S. Jo, T. Kim, V. G. Iyer, and W. Im. 38. J Comput Chem 29, 1859-1865 (2008).W. Im, M. S. Lee, and C. L. 3rd Brooks. 39. J Comput Chem 24, 1691-1702 (2003).K. Vanommeslaeghe, E. Hatcher, C. Acharya, S. Kundu, S. Zhong, J. Shim, E. Darian, 40. O. Guvench, P. Lopes, I. Vorobyov, and A. D. Mackerell, Jr. J Comput Chem 31, 671-690 (2010).M. X. Xie, X. Y. Xu, and Y. D. Wang. 41. Biochim Biophys Acta 1724, 215-224 (2005).P. Boukamp, R. T. Petrussevska, D. Breitkreutz, J. Hornung, A. Markham, and N. E. 42. Fusenig. J Cell Biol 106, 761-771 (1988).A. Sali and T. Blundell. 43. J Mol Biol 234, 779-815 (1993).I. B. Afanas’ev, E. A. Ostrachovich, and L. G. Korkina. 44. FEBS Lett 425, 256-258 (1998).Y. Bai, F. Song, M. Chen, J. Xing, Z. Liu, and S. Liu. 45. Anal Sci 20, 1147-1151 (2004).Y. D. Park, Y. J. Lyou, H. S. Hahn, M. J. Hahn, and J. M. Yang. 46. J Biomol Struct Dyn 24, (2006) 131-138.Y. D. Park, S. Y. Kim, Y. J. Lyou, D. Y. Lee, and J. M. Yang. 47. Biochem Cell Biol 84, 112-116 (2006).K. H. Park, J. R. Lee, H. S. Hahn, Y. H. Kim, C. D. Bae, J. M. Yang, S. Oh, Y. J. Bae, D. E. 48. Kim, and M. J. Hahn. Chem Pharm Bull (Tokyo) 54, 1266-1270 (2006).N. Gheibi, A. A. Saboury, H. Mansuri-Torshizi, K. Haghbeen, and A. A. Moosavi-Movahedi. 49. J Enzyme Inhib Med Chem 20, 393-399 (2005).S. Khatib, O. Nerya, R. Musa, M. Shmuel, S. Tamir, and J. Vaya. 50. Bioorg Med Chem 13, 433-441 (2005).Z. R. Lü, L. Shi, J. Wang, D. Park, J. Bhak, J. M. Yang, Y. D. Park, H. W. Zhou, and F. Zou. 51. Appl Biochem Biotechnol 160, 1896-908 (2010).Y. D. Park, J. Y. Jung, D. W. Kim, W. S. Kim, M. J. Hahn, and J. M. Yang. 52. J Protein Chem 22, 463-471 (2003).

Date Received: November 7, 2011

Communicated by the Editor Ramaswamy H. Sarma